Surging AI Chip Prices Pose New Risk to Tech Rally

The artificial intelligence boom faces a new economic challenge as semiconductor prices surge at unprecedented rates, according to a strategist at New York Life Investment Management.

Julia Hermann, global market strategist at the firm, has identified what she calls "chipflation"—the rapid escalation in prices for AI-related logic and memory chips—as a potential headwind that could test the durability of the AI investment thesis. The concern centers on whether companies pouring billions into AI infrastructure can sustain profitability as input costs climb.

Unprecedented price growth in memory chips

South Korea's export price index for DRAM memory chips illustrates the scale of the price surge. Hermann noted that in previous market cycles, memory chip price growth typically peaked around 100% year-over-year. Current data shows South Korean memory chip prices growing at 370% year-over-year—nearly four times historical highs.

The dramatic price increases reflect strong demand for AI computing capacity, but they also represent mounting costs for hyperscale cloud providers building out data center infrastructure. These companies face rising expenses not only from chips but also from energy and utility requirements, all while returns on their AI investments remain years away.

The double-edged nature of chip demand

Hermann described the situation as a balancing act with implications that cut both ways. While surging prices signal robust demand for AI capabilities, they simultaneously threaten to constrain the pace of infrastructure buildout by making expansion more expensive.

"Although the volatility associated with semiconductor leadership has been primarily to the upside this year, that volatility can work in both directions and test investor conviction," Hermann told Business Insider.

Chip stocks have experienced pressure in recent weeks as investors begin scrutinizing the economics underlying the AI rally. Major semiconductor stocks, including memory manufacturers and hardware providers, have declined despite largely beating earnings estimates in early reports.

Why it matters

The chipflation dynamic forces a reckoning between near-term costs and long-term AI potential. If component prices continue climbing at rates far exceeding historical norms, hyperscalers may need to slow infrastructure spending or accept compressed margins—either outcome could dampen enthusiasm for AI-related investments. The tension between strong demand signals and rising cost pressures will likely determine whether the current AI trade can sustain its momentum or requires a period of consolidation.

Hermann's investment approach in this environment prioritizes quality across the AI supply chain, focusing on companies with strong profitability, low-to-moderate earnings variability, and robust interest coverage—characteristics that may provide resilience if chipflation continues testing market conviction.

These details were first reported by Business Insider.